In the last decade the concept of developing anti-cancer medications which target abnormally active protein kinases has led to a number of successes. In addition to the actions of protein kinases, lipid kinases also play an important role in generating critical regulatory second messengers. The PI3K family of lipid kinases generates 3′-phosphoinositides that bind to and activate a variety of cellular targets, initiating a wide range of signal transduction cascades (Vanhaesebroeck et al., 2001; Toker, 2002; Pendaries et al., 2003; Downes et al., 2005). These cascades ultimately induce changes in multiple cellular processes, including cell proliferation, cell survival, differentiation, vesicle trafficking, migration, and chemotaxis.
PI3Ks can be divided into three distinct classes based upon differences in both structure, and substrate preference. While members of the Class II family of PI3Ks have been implicated in the regulation of tumor growth (Brown & Shepherd, 2001; Traer et al., 2006), the bulk of research has focused on the Class I enzymes and their role in cancer (Stauffer et al., 2005; Stephens et al., 2005; Vivanco & Sawyers, 2002; Workman, 2004; Chen et al., 2005; Hennessy et al., 2005; Cully et al., 2006). Class I PI3Ks have traditionally been divided into two distinct sub-classes based upon differences in protein subunit composition. The Class IA PI3Ks are comprised of a catalytic p110 catalytic subunit (p110α, β or δ) heterodimerized with a member of the p85 regulatory subunit family. In contrast, the Class IB PI3K catalytic subunit (p110γ) heterodimerizes with a distinct p101 regulatory subunit (reviewed by Vanhaesebroeck & Waterfield, 1999; Funaki et al., 2000; Katso et al., 2001). The C-terminal region of these proteins contains a catalytic domain that possesses distant homology to protein kinases. The PI3Kγ structure is similar to Class IA p110s, but lacks the N-terminal p85 binding site (Domin & Waterfield, 1997). Though similar in overall structure, the homology between catalytic p110 subunits is low to moderate. The highest homology between the PI3K isoforms is in the kinase pocket of the kinase domain.
The Class IA PI3K isoforms associate with activated receptor tyrosine kinases (RTKs) (including PDGFR, EGFR, VEGFR, IGF1-R, c-KIT, CSF-R and Met), or with tyrosine phosphorylated adapter proteins (such as Grb2, Cbl, IRS-1 or Gab1), via their p85 regulatory subunits resulting in stimulation of the lipid kinase activity. Activation of the lipid kinase activity of the p110β and p110γ isoforms has been shown to occur in response to binding to activated forms of the ras Oncogene (Kodaki et al., 1994). In fact, the oncogenic activity of these isoforms may require binding to ras (Kang et al., 2006). In contrast, the p110α and p110δ isoforms exhibit oncogenic activity independent of ras binding, through constitutive activation of Akt.
Class I PI3Ks catalyze the conversion of PI(4,5)P2 [PIP2] to PI(3,4,5)P3 [PIP3]. The production of PIP3 by PI3K affects multiple signaling processes that regulate and coordinate the biological end points of cell proliferation, cell survival, differentiation and cell migration. PIP3 is bound by Pleckstrin-Homology (PH) domain-containing proteins, including the phosphoinositide-dependent kinase, PDK1 and the Akt proto-oncogene product, localizing these proteins in regions of active signal transduction and also contributing directly to their activation (Klippel et al., 1997; Fleming et al., 2000; Itoh & Takenawa, 2002; Lemmon, 2003). This co-localization of PDK1 with Akt facilitates the phosphorylation and activation of Akt. Carboxy-terminal phosphorylation of Akt on Ser473 promotes phosphorylation of Thr308 in the Akt activation loop (Chan & Tsichlis, 2001; Hodgkinson et al., 2002; Scheid et al., 2002; Hresko et al., 2003). Once active, Akt phosphorylates and regulates multiple regulatory kinases of pathways that directly influence cell cycle progression and cell survival.
Many of the effects of Akt activation are mediated via its negative regulation of pathways which impact cell survival and which are commonly dysregulated in cancer. Akt promotes tumor cell survival by regulating components of the apoptotic and cell cycle machinery. Akt is one of several kinases that phosphorylate and inactivate pro-apoptotic BAD proteins (del Peso et al., 1997; Pastorino et al., 1999). Akt may also promote cell survival through blocking cytochrome C-dependent caspase activation by phosphorylating Caspase 9 on Ser196 (Cardone et al., 1998).
Akt impacts gene transcription on several levels. The Akt-mediated phosphorylation of the MDM2 E3 ubiquitin ligase on Ser166 and Ser186 facilitates the nuclear import of MDM2 and the formation and activation of the ubiquitin ligase complex. Nuclear MDM2 targets the p53 tumor suppressor for degradation, a process that can be blocked by LY294002 (Yap et al., 2000; Ogawara et al., 2002). Downregulation of p53 by MDM2 negatively impacts the transcription of p53-regulated pro-apoptotic genes (e.g. Bax, Fas, PUMA and DR5), the cell cycle inhibitor, p21Cip1, and the PTEN tumor suppressor (Momand et al., 2000; Hupp et al., 2000; Mayo et al., 2002; Su et al., 2003). Similarly, the Akt-mediated phosphorylation of the Forkhead transcription factors FKHR, FKHRL and AFX (Kops et al., 1999; Tang et al., 1999), facilitates their binding to 14-3-3 proteins and export from the cell nucleus to the cytosol (Brunet et al., 1999). This functional inactivation of Forkhead activity also impacts pro-apoptotic and pro-angiogenic gene transcription including the transcription of Fas ligand (Ciechomska et al., 2003) Bim, a pro-apoptotic Bcl-2 family member (Dijkers et al., 2000), and the Angiopoietin-1 (Ang-1) antagonist, Ang-2 (Daly et al., 2004). Forkhead transcription factors regulate the expression of the cyclin-dependent kinase (Cdk) inhibitor p27Kip1. Indeed, PI3K inhibitors have been demonstrated to induce p27Kip1 expression resulting in Cdk1 inhibition, cell cycle arrest and apoptosis (Dijkers et al., 2000). Akt is also reported to phosphorylate p21Cip1 on Thr145 and p27Kip1 on Thr157 facilitating their association with 14-3-3 proteins, resulting in nuclear export and cytoplasmic retention, preventing their inhibition of nuclear Cdks (Zhou et al., 2001; Motti et al., 2004; Sekimoto et al., 2004). In addition to these effects, Akt phosphorylates IKK (Romashkova & Makarov, 1999), leading to the phosphorylation and degradation of IκB and subsequent nuclear translocation of NFκB, resulting in the expression of survival genes such as IAP and Bcl-XL.
The PI3K/Akt pathway is also linked to the suppression of apoptosis through the JNK and p38MAPK MAP Kinases that are associated with the induction of apoptosis. Akt is postulated to suppress JNK and p38MAPK signaling through the phosphorylation and inhibition of two JNK/p38 regulatory kinases, Apoptosis Signal-regulating Kinase 1 (ASK1) (Kim et al., 2001; Liao & Hung, 2003; Yuan et al., 2003), and Mixed Lineage Kinase 3 (MLK3) (Lopez-Ilasaca et al., 1997; Barthwal et al., 2003; Figueroa et al., 2003). The induction of p38MAPK activity is observed in tumors treated with cytotoxic agents and is required for those agents to induce cell death (reviewed in Olson & Hallahan, 2004). Thus, inhibitors of the PI3K pathway may promote the activities of co-administered cytotoxic drugs.
An additional role for PI3K/Akt signaling involves the regulation of cell cycle progression through modulation of Glycogen Synthase Kinase 3 (GSK3) activity. GSK3 activity is elevated in quiescent cells, where it phosphorylates cyclin D1 on Ser286, targeting the protein for ubiquitination and degradation (Diehl et al., 1998) and blocking entry into S-phase. Akt inhibits GSK3 activity through phosphorylation on Ser9 (Cross et al., 1995). This results in the elevation of Cyclin D1 levels which promotes cell cycle progression. Inhibition of GSK3 activity also impacts cell proliferation through activation of the wnt/beta-catenin signaling pathway (Abbosh & Nephew, 2005; Naito et al., 2005; Wilker et al., 2005; Segrelles et al., 2006). Akt mediated phosphorylation of GSK3 results in stabilization and nuclear localization of the beta-catenin protein, which in turn leads to increased expression of c-myc and cyclin D1, targets of the beta-catenin/Tcf pathway.
Although PI3K signaling is utilized by many of the signal transduction networks associated with both oncogenes and tumor suppressors, PI3K and its activity have been linked directly to cancer. Overexpression of both the p110α and p110β isoforms has been observed in bladder and colon tumors and cell lines, and overexpression generally correlates with increased PI3K activity (Bénistant et al., 2000). Overexpression of p110α□□ has also been reported in ovarian and cervical tumors and tumor cell lines, as well as in squamous cell lung carcinomas. The overexpression of p110α in cervical and ovarian tumor lines is associated with increased PI3K activity (Shayesteh et al., 1999; Ma et al., 2000). Elevated PI3K activity has been observed in colorectal carcinomas (Phillips et al., 1998) and increased expression has been observed in breast carcinomas (Gershtein et al., 1999).
Over the last few years, somatic mutations in the gene encoding p110α (PIK3CA) have been identified in numerous cancers. The data collected to date suggests that PIK3CA is mutated in approximately 32% of colorectal cancers (Samuels et al., 2004; Ikenoue et al., 2005), 18-40% of breast cancers (Bachman et al., 2004; Campbell et al., 2004; Levine et al., 2005; Saal et al., 2005; Wu et al., 2005), 27% of glioblastomas (Samuels et al., 2004; Hartmann et al., 2005; Gallia et al., 2006), 25% of gastric cancers (Samuels et al., 2004; Byun et al., 2003; Li et al., 2005), 36% of hepatocellular carcinomas (Lee et al., 2005), 4-12% of ovarian cancers (Levine et al., 2005; Wang et al., 2005), 4% of lung cancers (Samuels et al., 2004; Whyte & Holbeck, 2006), and up to 40% of endometrial cancers (Oda et al., 2005). PIK3CA mutations have been reported in oligodendroma, astrocytoma, medulloblastoma, and thyroid tumors as well (Broderick et al., 2004; Garcia-Rostan et al., 2005). Based upon the observed high frequency of mutation, PIK3CA is one of the two most frequently mutated genes associated with cancer, the other being K-ras. More than 80% of the PIK3CA mutations cluster within two regions of the protein, the helical (E545K) and catalytic (H1047R) domains. Biochemical analysis and protein expression studies have demonstrated that both mutations lead to increased constitutive p110α catalytic activity and are in fact, oncogenic (Bader et al., 2006; Kang et al., 2005; Samuels et al., 2005; Samuels & Ericson, 2006). Recently, it has been reported that PIK3CA knockout mouse embryo fibroblasts are deficient in signaling downstream from various growth factor receptors (IGF-1, Insulin, PDGF, EGF), and are resistant to transformation by a variety of oncogenic RTKs (IGFR, wild-type EGFR and somatic activating mutants of EGFR, Her2/Neu) (Zhao et al., 2006).
Functional studies of PI3K in vivo have demonstrated that siRNA-mediated downregulation of p110β inhibits both Akt phosphorylation and HeLa cell tumor growth in nude mice (Czauderna et al., 2003). In similar experiments, siRNA-mediated downregulation of p110β was also shown to inhibit the growth of malignant glioma cells in vitro and in vivo (Pu et al., 2006). Inhibition of PI3K function by dominant-negative p85 regulatory subunits can block mitogenesis and cell transformation (Huang et al., 1996; Rahimi et al., 1996). Several somatic mutations in the genes encoding the p85α and p85β regulatory subunits of PI3K that result in elevated lipid kinase activity have been identified in a number of cancer cells as well (Janssen et al., 1998; Jimenez et al., 1998; Philp et al., 2001; Jucker et al., 2002; Shekar et al., 2005). Neutralizing PI3K antibodies also block mitogenesis and can induce apoptosis in vitro (Roche et al., 1994; Roche et al., 1998; Bénistant et al., 2000). In vivo proof-of-principle studies using the PI3K inhibitors LY294002 and wortmannin, demonstrate that inhibition of PI3K signaling slows tumor growth in vivo (Powis et al., 1994; Schultz et al., 1995; Semba et al., 2002; Ihle et al., 2004).
Overexpression of Class I PI3K activity, or stimulation of their lipid kinase activities, is associated with resistance to both targeted (such as imatinib and tratsuzumab) and cytotoxic chemotherapeutic approaches, as well as radiation therapy (West et al., 2002; Gupta et al., 2003; Osaki et al., 2004; Nagata et al., 2004; Gottschalk et al., 2005; Kim et al., 2005). Activation of PI3K has also been shown to lead to expression of multidrug resistant protein-1 (MRP-1) in prostate cancer cells and the subsequent induction of resistance to chemotherapy (Lee et al., 2004).
The importance of PI3K signaling in tumorigenesis is further underscored by the findings that the PTEN tumor suppressor, a PI(3)P phosphatase, is among the most commonly inactivated genes in human cancers (Li et al., 1997; Steck et al., 1997; Ali et al., 1999; Ishii et al., 1999). PTEN dephosphorylates PI(3,4,5)P3 to PI(4,5)P2 thereby antagonizing PI3K-dependent signaling. Cells containing functionally inactive PTEN have elevated levels of PIP3, high levels of activity of PI3K signaling (Haas-Kogan et al., 1998; Myers et al., 1998; Taylor et al., 2000), increased proliferative potential, and decreased sensitivity to pro-apoptotic stimuli (Stambolic et al., 1998). Reconstitution of a functional PTEN suppresses PI3K signaling (Taylor et al., 2000), inhibits cell growth and re-sensitizes cells to pro-apoptotic stimuli (Myers et al., 1998; Zhao et al., 2004). Similarly, restoration of PTEN function in tumors lacking functional PTEN inhibits tumor growth in vivo (Stahl et al., 2003; Su et al., 2003; Tanaka & Grossman, 2003) and sensitizes cells to cytotoxic agents (Tanaka & Grossman, 2003).
The signaling inputs to Class I PI3Ks are diverse and can be deduced through genetic analyses. Thus, activation of AKT was impaired in p110α-deficient murine embryonic fibroblasts (MEFs) upon stimulation by classical Receptor Tyrosine Kinase (RTK) ligands (e.g., EGF, insulin, IGF-1, and PDGF) (Zhao et al., 2006). However, MEFs in which p110β is ablated or replaced by a kinase-dead allele of p110β respond normally to growth factor stimulation via RTKs (Jia et al., 2008). In contrast, p110β catalytic activity is required for AKT activation in response to GPCR ligands (such as LPA). As such, p110α appears to carry the majority of the PI3K signal in classic RTK signaling and is responsible for tumor cell growth, proliferation, survival, angiogenesis and metabolism, whereas p110β mediates GPCR signaling from mitogens and chemokines and therefore may regulate tumor cell proliferation, metabolism, inflammation and invasion (Vogt et al., 2009; Jia et al., 2009).
The mutation of the gene encoding p110β is rare in tumors, but amplification of PI3Kβ has been found in many tumors (Bénistant et al., 2000; Brugge et al., 2007). Importantly, in a mouse prostate tumor model driven by PTEN deficiency, ablation of p110α was shown to have no effect on tumorigenesis (Jia et al., 2008). Furthermore, in PTEN-deficient human cancer cell lines (e.g., PC-3, U87MG, and BT549) of p110β, but not p110α, inhibits downstream activation of AKT, cell transformation, and the growth of PTEN-deficient cells and tumor xenografts (Wee et al., 2008). Genetic studies have suggested that the kinase activity of p110β is essential in cellular transformation caused by PTEN loss. For example, adding back a kinase-dead p110β, but not its wild-type counterpart, impaired focus formation in PTEN-deficient PC3 cells depleted for endogenous p110β (Wee et al., 2008). These studies demonstrate that PTEN-deficient tumor cells depend on p110β and its catalytic activity for signaling and growth.
Genetic alteration of tumor suppressor gene PTEN is frequently found in many cancers (Liu et al., 2009), such as endometrial cancer (43%), CRPC (35-79%), glioma (19%) and melanoma (18%). In the case of endometrial cancer, coexisting PIK3CA and PTEN genetic alteration was confirmed (Yuan & Cantley, 2008). In addition to mutation, amplification of PIK3CA and loss-of-function of PTEN by various molecular mechanisms have been discovered. For example, amplification of PIK3CA and loss-of-function of PTEN was found in 30-50% and 35-60% of gastric cancer patients, respectively, although PIK3CA and PTEN mutation rate was reported to be less than 7% of each (Byun et al., 2003; Oki et al., 2006; Li et al., 2005; Sanger Database).
While a subset of tumor types are solely dependent on PI3Kα signaling, other tumors are dependent on PI3Kβ signaling or on a combination of both PI3Kα and PI3Kβ signaling.
Therefore, there remains a need for balanced PI3K α/β inhibitors capable of inhibiting both PI3K alpha and beta targets.
WO 2008/070150 (Bayer Schering Pharma Aktiengesellschaft) relates to 2,3-dihydroimidazo[1,2-c]quinazoline compounds, to pharmaceutical compositions containing such compounds and the use of such compounds or compositions for phosphotidylinositol-3-kinase (PI3K) inhibition, and treating diseases associated with PI3K activity, in particular treating hyper-proliferative and/or angiogenesis disorders, as a sole agent or in combination with other active ingredients. Said compounds show an increased activity (lower IC50) against PI3K alpha than against PI3k beta.
However, the state of the art described above does not describe the compounds of general formula (I) of the present invention, a stereoisomer, a tautomer, an N-oxide, a hydrate, a solvate, or a salt thereof, or a mixture of same, as described and defined in the claims herein, and as hereinafter referred to as “compounds of the present invention”. Nor does the state of the art described above show the pharmacological activity as shown by the compounds of general formula (I) of the present invention.
It has now been found, and this constitutes the basis of the present invention, that said compounds of the present invention, as described and defined herein, and as hereinafter referred to as “compounds of the present invention”, have surprising and advantageous properties: the compounds of the present invention display surprising balanced activity for the inhibition of phosphatidylinositol-3-kinase alpha- and beta-isoforms as shown in the biologiocal section of this text, which is shown as the ratio PI3K beta IC50/PI3K alpha IC50.
The compounds of the present invention, including salts, metabolites, solvates, solvates of salts, hydrates, and stereoisomeric forms thereof, exhibit anti-proliferative activity and are thus useful to prevent or treat the disorders associated with hyper-proliferation: in particular, said compounds of general formula (I) of the present invention may therefore be used for the treatment or prophylaxis of diseases of uncontrolled cell growth, proliferation and/or survival, inappropriate cellular immune responses, or inappropriate cellular inflammatory responses or diseases which are accompanied with uncontrolled cell growth, proliferation and/or survival, inappropriate cellular immune responses, or inappropriate cellular inflammatory responses, particularly in which the uncontrolled cell growth, proliferation and/or survival, inappropriate cellular immune responses, or inappropriate cellular inflammatory responses is mediated by the PI3K pathway, such as, for example, haemotological tumours, solid tumours, and/or metastases thereof, e.g. leukaemias and myelodysplastic syndrome, malignant lymphomas, head and neck tumours including brain tumours and brain metastases, tumours of the thorax including non-small cell and small cell lung tumours, gastrointestinal tumours, endocrine tumours, mammary and other gynaecological tumours, urological tumours including renal, bladder and prostate tumours, skin tumours, and sarcomas, and/or metastases thereof.